COMMENTS. Confirming More Guns, Less Crime. Florenz Plassmann* & John Whitley**

Transcription

1 COMMENTS Confirming More Guns, Less Crime Florenz Plassmann* & John Whitley** Analyzing county-level data for the entire United States from 1977 to 2000, we find annual reductions in murder rates between 1.5% and 2.3% for each additional year that a right-to-carry law is in effect. For the first five years that such a law is in effect, the total benefit from reduced crimes usually ranges between approximately $2 billion and $3 billion per year. Ayres and Donohue have simply misread their own results. Their own most general specification that breaks down the impact of the law on a year-byyear basis shows large crime-reducing benefits. Virtually none of their claims that their county-level hybrid model implies initial significant increases in crime are correct. Overall, the vast majority of their estimates based on data up to 1997 actually demonstrate that right-to-carry laws produce substantial crime-reducing benefits. We show that their models also do an extremely poor job of predicting the changes in crime rates after INTRODUCTION I. WHAT DOES THEIR EVIDENCE SHOW? A. Is There a Robbery Effect? B. Murder Rates C. Rapes and Aggravated Assaults D. Critiques of Year-by-Year Breakdown of the Law s Impact E. The Estimated Benefits from the Law II. COUNTY-LEVEL DATA FROM 1977 TO A. Advantages and Disadvantages of Different Types of Data B. Extending the Data to C. Is the Adoption of a Right-to-Carry Law Endogenous? D. Poisson Estimates III. EVALUATING SOME GENERAL CLAIMS MADE BY AYRES AND DONOHUE * Department of Economics, State University of New York at Binghamton. ** School of Economics, University of Adelaide. We thank John Lott for his support, comments and discussion. 1313

2 1314 STANFORD LAW REVIEW [Vol. 55:1313 A. Has Previous Work Acknowledged Both the Costs and Benefits of Guns? B. Possible Bad Effects of Concealed Handgun Laws C. Other Concerns: The Risks to Police, Accidents, and the Problem with Untrained Permit Holders D. Are There Initial Jumps in Crime? E. Can Cocaine Use Explain the Results? F. Measurement Error in County-Level Data G. Is the U.S. Murder Rate Exceptional? H. Should Philadelphia Be Treated Differently from the Rest of Pennsylvania After 1989? CONCLUSION FIGURES FIGURE 1A: AYRES AND DONOHUE S ESTIMATED IMPACTS ON ROBBERY (ALL THE YEAR-BY-YEAR ESTIMATES THAT IAN AYRES AND JOHN DONOHUE REPORT) FIGURE 1B: AYRES AND DONOHUE S ESTIMATED IMPACTS ON MURDER (ALL THE YEAR-BY-YEAR ESTIMATES THAT IAN AYRES AND JOHN DONOHUE REPORT) FIGURE 1C: AYRES AND DONOHUE S ESTIMATED IMPACTS ON RAPE (ALL THE YEAR-BY-YEAR ESTIMATES THAT IAN AYRES AND JOHN DONOHUE REPORT) FIGURE 1D: AYRES AND DONOHUE S ESTIMATED IMPACTS ON AGGRAVATED ASSAULT (ALL THE YEAR-BY-YEAR ESTIMATES THAT IAN AYRES AND JOHN DONOHUE REPORT) FIGURE 2: SHOWING WHAT HAPPENS WHEN YOU FIT AN INTERCEPT SHIFT AND A TREND TO A NONLINEAR TREND FIGURE 3: EXAMINING THE LAST STATES TO ADOPT THE RIGHT-TO-CARRY LAW FIGURE 4A: VIOLENT CRIME: WEIGHTED LEAST SQUARES ESTIMATES USING COUNTY-LEVEL DATA FROM 1977 TO FIGURE 4B: MURDER: WEIGHTED LEAST SQUARES ESTIMATES USING COUNTY-LEVEL DATA FROM 1977 TO FIGURE 4C: RAPE: WEIGHTED LEAST SQUARES ESTIMATES USING COUNTY- LEVEL DATA FROM 1977 TO FIGURE 4D: ROBBERY: WEIGHTED LEAST SQUARES ESTIMATES USING COUNTY-LEVEL DATA FROM 1977 TO FIGURE 4E: AGGRAVATED ASSAULT: WEIGHTED LEAST SQUARES ESTIMATES USING COUNTY-LEVEL DATA FROM 1977 TO FIGURE 4F: PROPERTY CRIMES: WEIGHTED LEAST SQUARES ESTIMATES USING COUNTY-LEVEL DATA FROM 1977 TO

3 Apr. 2003] CONFIRMING MORE GUNS, LESS CRIME 1315 TABLES TABLE 1: COST-BENEFIT ANALYSES OF AYRES AND DONOHUE S COUNTY-LEVEL REGRESSIONS: FIVE-YEAR AVERAGE ESTIMATE OF NET COSTS/BENEFITS TABLE 2: COST-BENEFIT ANALYSES OF COUNTY- AND STATE-LEVEL REGRESSIONS IN DONOHUE S BROOKINGS PAPER: FIVE-YEAR AVERAGE ESTIMATE OF NET COSTS/BENEFITS TABLE 3A: COMPARISON OF THE THREE DUMMY SPECIFICATIONS USED BY AYRES AND DONOHUE (OUR MODEL INCLUDES REGION-SPECIFIC YEAR DUMMIES AND USES DATA) TABLE 3B: LIMITED SET OF DEMOGRAPHICS AND LAGGED PER CAPITA PRISON POPULATION, DATA TABLE 4: EARLY AND LATE ADOPTERS, FULL SET OF DEMOGRAPHICS, DATA TABLE 5: TEST WHETHER THE ADOPTION OF SHALL-ISSUE LAWS IS ENDOGENOUS TO THE CRIME RATE TABLE 6: STATE-BY-STATE BREAKDOWN: SPLINE MODEL, COUNTY-LEVEL DATA, WITH REGIONAL YEAR FIXED EFFECTS TABLE 7: POISSON ESTIMATES: ALL SOCIOECONOMIC VARIABLES AND THE ARREST RATE OF VIOLENT CRIME TABLE 8: ALL SOCIOECONOMIC VARIABLES AND THE ARREST RATE OF VIOLENT CRIME APPENDIX TABLE 1: REPORTING THE RESULTS ON VIOLENT CRIME RATES FROM STUDIES CRITICAL OF RIGHT-TO-CARRY LAWS (USING THE NATIONAL COEFFICIENTS FROM THE MOST CRITICAL STUDIES THAT EXAMINED THE CHANGE IN CRIME RATES BEFORE-AND-AFTER THE PASSAGE OF RIGHT-TO-CARRY LAWS) APPENDIX TABLE 2: ANALYSIS WITH YEAR-SPECIFIC TREND DUMMIES (OUR MODEL INCLUDES REGION-SPECIFIC YEAR DUMMIES AND USES DATA) APPENDIX TABLE 3: LIMITED DEMOGRAPHIC VARIABLES AND ADDING LAGGED PRISON POPULATION, APPENDIX TABLE 4: SPLINE MODEL, STATE-LEVEL DATA, REGIONAL YEAR FIXED EFFECTS INTRODUCTION Quite a few empirical papers have examined the impact of right-to-carry laws on crime rates. Most studies have found significant benefits, with some finding reductions in murder rates twice as large as the original research For a summary, see John R. Lott, Jr., Guns, Crime, and Safety: Introduction, 44 J.L. & ECON. 605, (2001). Individual papers that show a benefit from the law include William Alan Bartley & Mark A. Cohen, The Effect of Concealed Weapons Laws: An Extreme Bound Analysis, 36 ECON. INQUIRY 258, (1998); Stephen G. Bronars & John R. Lott, Jr., Criminal Deterrence, Geographic Spillovers, and Right-to-Carry Laws, AM.

4 1316 STANFORD LAW REVIEW [Vol. 55:1313 Even the critics did not provide evidence that such laws have increased violent crime, accidental gun deaths, or suicides. 2 Unlike previous authors, Ian Ayres and John Donohue claim to have found significant evidence that right-to-carry laws increased crime. However, they have misread their own results. The most detailed results they report following the change in crime rates on a year-by-year basis before and after right-to-carry laws were adopted clearly show large drops in violent crime that occur immediately after the laws were adopted. Their hybrid results showing a small increase in crime immediately after passage are not statistically significant and are an artifact of fitting a straight line to a curved one. When one examines a longer period from 1977 to 2000 even this type of result disappears. Ayres and Donohue s efforts have been valuable in forcing others to reexamine the evidence, extend the dataset over more years, and think of new ways to test hypotheses, and we appreciate their efforts. 3 They are both highly regarded and well-known for their research, such as claiming that the legalization of abortion can account for half the drop in murder during the 1990s. 4 Unfortunately, their research on this issue inaccurately describes the literature and also fails to address previous critiques of their work. For example, Ayres and Donohue claim that [w]hen we added five years of county data and seven years of state data, allowing us to test an additional fourteen jurisdictions that adopted shall-issue laws, the previous Lott and Mustard findings proved not to be robust. 5 All their tables report results for Lott s ECON. REV., May 1998, at ; John R. Lott, Jr. & John E. Whitley, Safe-Storage Gun Laws: Accidental Deaths, Suicides, and Crime, 44 J.L. & ECON. 659, (2001); Tomas B. Marvell, The Impact of Banning Juvenile Gun Possession, 44 J.L. & ECON. 691, (2001); Carlisle E. Moody, Testing for the Effects of Concealed Weapons Laws: Specification Errors and Robustness, 44 J.L. & ECON. 799, (2001); David B. Mustard, The Impact of Gun Laws on Police Deaths, 44 J.L. & ECON. 635, (2001); David E. Olson & Michael D. Maltz, Right-to-Carry Concealed Weapon Laws and Homicide in Large U.S. Counties: The Effect on Weapon Types, Victim Characteristics, and Victim-Offender Relationships, 44 J.L. & ECON. 747, (2001); Florenz Plassmann & T. Nicolaus Tideman, Does the Right to Carry Concealed Handguns Deter Countable Crimes? Only a Count Analysis Can Say, 44 J.L. & ECON. 771, (2001); Eric Helland & Alexander Tabarrok, Using Placebo Laws to Test More Guns, Less Crime : A Note (Univ. of Chi. Graduate Sch. of Bus., Working Paper, 2002). 2. In fact, these critics provided a great deal of supportive evidence. See infra app. tbl For an earlier discussion on Lott s research, see Ian Ayres & John J. Donohue III, Nondiscretionary Concealed Weapons Law: A Case Study of Statistics, Standards of Proof, and Public Policy, 1 AM. L. & ECON. REV. 436 (1999); John R. Lott, Jr., More Guns, Less Crime: A Response to Ayres and Donohue (Yale Law Sch., Working Paper, 1999), available at 20Download. 4. John J. Donohue III & Steven D. Levitt, The Impact of Legalizing Abortion on Crime Rates, 116 Q.J. ECON. 379, (2001). 5. Ian Ayres & John J. Donohue III, Shooting Down the More Guns, Less Crime

5 Apr. 2003] CONFIRMING MORE GUNS, LESS CRIME 1317 Time Period ( ) and compare those estimates with the Entire Period ( ). Yet, whatever differences in results arise, they are not due to the inclusion of more data for a longer period. Their paper gives a misleading impression as to how much their research extends the data period, since Lott s book and other work examined both the county and state data up through Ayres and Donohue s work thus extends the county-level data by one year, from twenty to twenty-one years. Part I of this Response reviews some of Ayres and Donohue s claims and shows that even their own estimates imply fairly consistently large annual benefits from reducing crime. We then extend the U.S. county-level data to 2000 in Part II, and, consistent with previous work, find large benefits from states adopting right-to-carry laws. As others have already found, the results are not sensitive to the inclusions of particular control variables, such as demographic measures. Finally, Part III provides direct responses and corrections to several specific claims made by Ayres and Donohue. I. WHAT DOES THEIR EVIDENCE SHOW? The most general specifications show the year-by-year changes in crime rates before and after the enactment of a right-to-carry law. Ayres and Donohue provide this breakdown for state-level data from 1977 to Their state-level data show the crime rates in the first year after the law was passed, the second year, and so on. While we disagree with some of their assumptions, their results provide a very useful starting point as their results stake out one side of the debate. John Donohue has another paper addressing these issues published by the Brookings Institution. 7 This paper presents the year-by-year changes for county-level data from 1977 to The county-level estimates report the crime rates in two-year intervals, and a separate dummy variable measures the combined effects whenever the state has had the law for eight or more years. 8 Hypothesis, 55 STAN. L. REV. 1193, 1296 (2003). 6. In one footnote Ayres and Donohue acknowledge that these additional data were used, but they claim that Lott only reports results for this dataset from tests of the trend specification. Id. at 1265 n.90. In Lott s book, figures 9.1 to 9.5 provide information on the nonlinear before-and-after trends; table 9.3 reports the relationship between the percent of the population with permits and crime rates (both linearly and nonlinearly); figures 9.6 to 9.9 show the impact of interacting the percent of the adult population with permits to different demographic characteristics; and figures 9.10 to 9.13 examine the sensitivity of the estimates to different combinations of the control variables. See JOHN R. LOTT, JR., MORE GUNS, LESS CRIME (2d ed. 2000). Other work that has examined the data through 1996 includes Lott & Whitley, supra note 1, and Mustard, supra note John J. Donohue, The Impact of Concealed-Carry Laws, in EVALUATING GUN POLICY: EFFECTS ON CRIME AND VIOLENCE 287 (Philip J. Cook & Jens Ludwig eds., 2003). 8. There are several advantages to the approach used for their county-level estimates. Using the two-year interval approach provides a better measure of trends without the

6 1318 STANFORD LAW REVIEW [Vol. 55:1313 One of the county estimates includes a separate state time trend for each state. Although our principal focus is on Ayres and Donohue s joint paper, we will refer to Donohue s Brookings paper several times in this Part because the papers are very similar and because several important results (like the countylevel year-by-year breakdowns) are only reported in the Brookings paper. Note also that the county and state estimates use two different definitions of the implementation of state right-to-carry laws, with the county-level data using a corrected version of the dates that Lott and Mustard used from Kopel and Cramer, and the state-level data using definitions supplied by Vernick and Hepburn. 9 A. Is There a Robbery Effect? Robbery is a good place to start our inquiry because it is committed in public more than any other crime, and should be the crime most likely to decline if the Lott and Mustard story of deterrence has any plausibility. 10 [T]he failure of the model to show a drop in robbery[] cast[s] doubt on the causal story that they advance. 11 Ayres and Donohue have consistently argued in several papers that robbery is the key result upon which the deterrence by right-to-carry laws is based. 12 In contrast, Lott has argued many times that there is no a priori reason to believe that the benefits of right-to-carry laws are larger for robbery than for other violent crimes. 13 Putting that debate aside, the robbery results presented by constraint of making all years have the same trend. The wild swings in both directions, as at the thirteenth and fourteenth year points in their state-level data, is simply due to only Maine being present for those observations. Examining just one state a decade and a half or more after a law is passed poses problems, particularly with state-level data where there is only one observation per year. Not only does it raise questions about what other factors may have changed in just that one state, but it also leads to extremely large confidence intervals. Thus, very little weight should be placed on those estimates, whether they are decreasing or increasing. 9. Jon Vernick & Lisa Hepburn, Description and Analysis of State and Federal Laws Affecting Firearm Manufacture, Sale, Possession, and Use, (Johns Hopkins Ctr. for Gun Policy Research, Working Paper, 2001); see also infra tbl Ayres & Donohue, supra note 5, at Id. at Ayres & Donohue, supra note 3; see also Ayres & Donohue, supra note 5, at 1206, 1227 (noting that robbery is an important result in determining the deterrent effect of rightto-carry laws). 13. Lott notes that residential robberies are the second largest category of robbery and that concealed handgun laws could actually cause them to rise as criminals substitute out of street robberies. See LOTT, supra note 6, at Just as criminals may switch between robbery and burglary, Lott also notes, but to rank some of these different crimes [murder, rape, robbery, and aggravated assault], one requires information on how sensitive different types of criminals are to the increased threat. Id. at 134.

7 Apr. 2003] CONFIRMING MORE GUNS, LESS CRIME 1319 Ayres and Donohue present a very clear, consistent story. 14 The state-level analysis shows that robbery rates continued rising, though at a slower rate, for the first two years after the law was passed. After that, however, robbery rates in right-to-carry states fell relative to non-right-to-carry states for the next nine years, and then remained fairly constant through year seventeen. The two sets of county-level estimates are even more dramatic. Robbery rates in right-tocarry states were rising until the laws were passed and then fell continually after that point. The pattern is very similar to that shown earlier by Lott in examining county-level data from 1977 to The changes are also very large. By the time the law has been in effect for six years, the county and state-level data imply a drop in robbery rates of eight and twelve percent respectively. It is difficult to see how anyone could look at these year-by-year results and accept their claims that the robbery effect is sensitive to the time frame examined or to the coding of when state laws were adopted. While Ayres and Donohue acknowledge the problems in using simple before-and-after averages in evaluating the impact of the law, they do not consistently apply that insight when discussing the evidence. 14. See infra fig.1a. 15. See LOTT, supra note 6, at

8 Estimated Impact on Crime FIGURE 1A: AYRES AND DONOHUE S ESTIMATED IMPACTS ON ROBBERY (ALL THE YEAR-BY-YEAR ESTIMATES THAT IAN AYRES AND JOHN DONOHUE REPORT) Years Before-and-After the Passage of the Right-to-Carry Laws Results from John Donohue, County Results, Table 5 Results from John Donohue, County Results with Individual State Trends, Table 7 Results from Ian Ayres and John Donohue, Figure 3e STANFORD LAW REVIEW [Vol. 55:1313 PLASSMAN FIGURE 1A 4/16/2003 6:25 PM

9 Apr. 2003] CONFIRMING MORE GUNS, LESS CRIME 1321 B. Murder Rates Figure 1b illustrates Ayres and Donohue s own year-by-year estimates for murder. Their county and state estimates paint a very consistent picture, but they dismiss the fact that state data estimated a 4.5% drop in murder rates during the first three years of the law as showing relatively little movement. 16 Their state-level regressions indicate that murder rates were rising in the three years prior to the law being passed and then falling over the next thirteen years. Only one state, Maine, has had the law in effect for more than thirteen years. The increase during years fourteen, fifteen, sixteen, and seventeen thus solely reflects changes in Maine s murder rate, and since this is state-level data, each coefficient represents only one data point. The values for these four years show up in the data only because Ayres and Donohue recode Maine s right-to-carry law as going into effect in 1981 instead of 1985 as previous research had done. 17 The increase between years thirteen and fourteen is also more apparent than real. The real increase is actually not due to any sudden change in Maine s crime rates, but to the fact that other states are included in calculating the crime rate for year thirteen, while only Maine is used for year fourteen. Both sets of county-level data again imply a large drop in crime that begins immediately after the law has been adopted and continues sharply down after that point. 18 By the time the law has been in effect for six years, Ayres and Donohue s very own county and state estimates imply that murder rates had fallen by at least ten percent. 16. Ayres & Donohue, supra note 5, at The laws in 1981 and 1985 differed in one crucial aspect: Under the 1981 law, city councils and mayors had responsibility for issuing permits. However, the police chiefs in Portland (with almost 20% of the state s population in 1985) and other major cities resisted issuing permits. The 1985 law overcame this problem by taking this power away from the city governments, particularly the Portland police chief, Frank Maoroso. Interview with Randy Kozuch, Legislative Affairs Director, NRA (Fall 2002). 18. Ayres and Donohue lump together the year of passage (year zero) with the first full year that the law is in effect, but if they had separated out the two, both the murder and robbery results would have also shown a drop in crimes between these two years.

10 Estimated Impact on Crime FIGURE 1B: AYRES AND DONOHUE S ESTIMATED IMPACTS ON MURDER (ALL THE YEAR-BY-YEAR ESTIMATES THAT IAN AYRES AND JOHN DONOHUE REPORT) Years Before-and-After the Passage of the Right-to-Carry Laws Results from John Donohue, County Results, Table 5 Results from John Donohue, County Results with Individual State Trends, Table 7 Results from Ian Ayres and John Donohue, Figure 3b STANFORD LAW REVIEW [Vol. 55:1313 PLASSMAN FIGURE 1B 4/16/2003 6:26 PM

11 Apr. 2003] CONFIRMING MORE GUNS, LESS CRIME 1323 C. Rapes and Aggravated Assaults Ayres and Donohue s county and state-level results for rapes and aggravated assaults are more ambiguous. The county-level estimates without the individual state trends show that both rape and aggravated assault fell almost continually after the laws were enacted. 19 Even choosing for comparison the sixth year after the law, where rape and aggravated assault rates have slightly risen back up, this still leaves rapes nine percent below their peak and aggravated assaults three percent below theirs. The county-level estimates with individual state trends provide a mixed picture. With the exception of one single year, rape rates rise before the law and fall thereafter. In stark contrast, using individual state trends changes the aggravated assault rate into a line that rises continuously over almost the entire period until the law has been in effect for eight or more years. Yet, since the aggravated assault rate was rising for years prior to the law at least as fast as it was rising after the law was passed, it is hard to blame the right-to-carry law for this rise. Ayres and Donohue s state-level results are also somewhat ambiguous, though even here the rape rates fall by ten percent for the first six years after the adoption of the law, and remain below the prelaw levels for at least twelve years. Only when Maine becomes the sole remaining state in the sample does the rape rate rise, and it rises above the prelaw levels for just one year (by seven percent). The rape rate then plunges by over twenty-five percent. With only one crime observation present here, the confidence intervals are so large that even with these wild swings, the changes are too small to conclude that the temporary surge in rapes placed it above the prelaw levels. There is only one year out of the seventeen years after the law was passed that the rape rate exceeds any of the values during the twelve years before the law. This is similar for aggravated assaults: Only three of the seventeen years after the adoption of the law show higher rates than any of the ten years prior to the law. 19. See infra figs.1c, 1d.

12 Estimated Impact on Crime FIGURE 1C: AYRES AND DONOHUE S ESTIMATED IMPACTS ON RAPE (ALL THE YEAR-BY-YEAR ESTIMATES THAT IAN AYRES AND JOHN DONOHUE REPORT) Years Before-and-After the Passage of the Right-to-Carry Laws Results from John Donohue, County Results, Table 5 Results from John Donohue, County Results with Individual State Trends, Table 7 Results from Ian Ayres and John Donohue, Figure 3c STANFORD LAW REVIEW [Vol. 55:1313 PLASSMAN FIGURE 1C 4/16/2003 6:26 PM

13 Estimated Impact on Crime FIGURE 1D: AYRES AND DONOHUE S ESTIMATED IMPACTS ON AGGRAVATED ASSAULT (ALL THE YEAR-BY-YEAR ESTIMATES THAT IAN AYRES AND JOHN DONOHUE REPORT) Years Before-and-After the Passage of the Right-to-Carry Laws Results from John Donohue, County Results, Table 5 Results from John Donohue, County Results with Individual State Trends, Table 7 Results from Ian Ayres and John Donohue, Figure 3d. Apr. 2003] CONFIRMING MORE GUNS, LESS CRIME 1325 PLASSMAN FIGURE 1D 4/16/2003 6:26 PM

14 1326 STANFORD LAW REVIEW [Vol. 55:1313 D. Critiques of Year-by-Year Breakdown of the Law s Impact The debate over simple dummies, splines, or hybrids becomes irrelevant when one has examined the year-by-year breakdown. All these approaches are simple ways to summarize the crime patterns and can provide useful statistics to test whether there is a change in crime rates, but the year-by-year dummies provide a much more accurate picture of changing crime patterns. It is important to understand what these year-by-year dummy estimates represent. The question is not whether these coefficients are different from zero, 20 but whether they have changed relative to other coefficients. A positive coefficient shows that the crime rate was higher in right-to-carry states than in non-right-to-carry states for that year, and coefficients that become less positive over time indicate that the crime rates in right-to-carry states are declining relative to other states. The patterns for robbery, murder, and rape clearly show that the longer the law is in effect, the greater the drop in crime. It is also not relevant, as Ayres and Donohue suggest, to compare the crime levels before and after the law. 21 When crime rates rise before the law and fall afterwards, there might be very little change in the before-and-after means even though, as the diagram for robbery indicates, something dramatic has changed. The key is to compare the trends before and after the law, and Ayres and Donohue s results in tables 10 and 11 imply large and statistically significant changes. 22 The year-by-year results do not support their claim that the main effect of the shall-issue laws is positive but over time this effect gets overwhelmed as the linear trend turns crime down. 23 Their county-level results indicate that by 20. Donohue advocates this interpretation when he writes: For the period the effect for the two or three years after dummy is seen to be highly positive and statistically significant in seven of the nine categories. The other two categories are insignificant, with one negative (murder) and one positive (rape). Donohue, supra note 7, at See Ayres & Donohue, supra note 5, at A comparison of years 1 and 2 before the law with years 2 and 3 after the law shows consistent declines. Murder declines from 2.9 to 4.2, rape from 3.7 to 2.6, robbery from 13.2 to 11, and aggravated assault from 6.7 to 6.3. These differences are not statistically significant by themselves, but as part of the trends they represent, the before-and-after trends are statistically significantly different from each other. 22. In Donohue s Brookings paper he writes: Lott mentions... the so-called inverted V hypothesis. Although there might be some hint of this..., the effects are not statistically significant (and, even if real, could be caused by a regression to the mean effect as opposed to a benign influence of the shall-issue law). Donohue, supra note 7, at 312. To test this, one must compare the before-and-after trends, and Ayres and Donohue s own estimates in their current paper do not support this claim. Both the spline and hybrid estimates on the 1977 to 1997 data reported in tables 10 and 11 of the Ayres and Donohue paper indicate consistent statistically significant changes in trends. 23. Ayres & Donohue, supra note 5, at 1264.

15 Apr. 2003] CONFIRMING MORE GUNS, LESS CRIME 1327 the second and third years after the law has been adopted, all violent crime rates are below the values that they had been in the two years preceding passage of the law. The figures and the standard errors associated with these estimates also allow us to directly evaluate their claims of model misspecification. One concern is about the claim regarding the county-level data that Donohue makes in his Brookings paper: This particular result of a positive main effect and a negative trend effect is inconsistent with any plausible theoretical prediction of the impact of a shall-issue law, since it is not clear why the law should initially accelerate crime and then dampen it. 24 Yet, Ayres and Donohue s year-byyear estimates shown in our Figures 1a through 1d indicate that no such positive main effect is occurring. A claim might be made that the hybrid model is misspecified solely because Ayres and Donohue are fitting a straight line to a nonlinear relationship. As an example, take Figure 2, where the crime rate is falling at an increasing rate after the adoption of the right-to-carry law. In order to fit a regression with both an intercept shift and a linear trend line to these nonlinear data, the intercept will have to be positive and the trend line will become steeper compared to a specification that uses only a trend line but no intercept shift. 25 This does not mean that there is actually an initial increase in crime, but only that it is an artifact of fitting a straight line to nonlinear data. Even a quadratic curve with an intercept may overestimate the crime rates in the initial years. The only solution is to measure the changes in crime on a year-by-year basis. 24. Donohue, supra note 7, at 297. Ayres and Donohue, supra note 5, at 1264, write: [T]he main effect of the shall-issue laws is positive but over time this effect gets overwhelmed.... See also id. at Indeed, if one compares the spline and the hybrid estimates in all of Ayres and Donohue s tables that compare the spline and the hybrid models, this is exactly the pattern that one observes. Comparison of the results in lines 5 and 6 of these tables shows that the positive intercept shift is associated with the trend line becoming more steeply negative.

16 1328 STANFORD LAW REVIEW [Vol. 55:1313 FIGURE 2: SHOWING WHAT HAPPENS WHEN YOU FIT AN INTERCEPT SHIFT AND A TREND TO A NONLINEAR TREND Intercept for after trend Crime Rate { Estimated crime trend using an intercept for the after trend as well as an after trend Actual data trend Estimated crime trend using just a trend line Years Before-and-After Adoption of the Law We can use their tables to address a second type of misinterpretation of estimation results. In discussing the state-level year-by-year estimates shown in their figures 3a through 3i, Ayres and Donohue note that: [O]ne can see that in four of the five violent crime categories and for burglary, even before adopting states passed their shall-issue legislation, crime was substantially higher than the regression model would have otherwise predicted (given the full array of explanatory variables). This raises concerns about the reliability of the regression model Their statement that crime was substantially higher is misleading because the differences are not statistically significant. While Ayres and Donohue s statelevel figures 3a through 3i do not report standard errors, this information is reported in Donohue s Brookings paper for the 1977 to 1997 time frame. 27 The crime rates for violent crimes, rape, robbery, and aggravated assault were never 26. Ayres & Donohue, supra note 5, at Donohue, supra note 7, at tbl.8-6. The differences in point estimate values between Donohue s table 8-6, id., and Ayres and Donohue s figures 3a-3i, Ayres & Donohue, supra note 5, at , simply arise because Ayres and Donohue include all possible year dummies for their figures, while only a portion of them is included for Donohue s Brookings paper table 8-6. The addition of two years to the dataset is not the crucial difference. The difference is that in Donohue s Brookings paper s table 6, the year values prior to year 6 are in the intercept term. Raising the intercept term reduces the size of the coefficients for the remaining year-by-year dummies. The relative pattern of the yearby-year dummies remains unchanged, but their significance relative to zero does change. This very point makes it clear how arbitrary it is to focus on whether these year-by-year dummies are different from zero, and not the more relevant question of whether the year-byyear dummies differ from each other in systematic ways.

17 Apr. 2003] CONFIRMING MORE GUNS, LESS CRIME 1329 statistically significantly different from zero (we assume that Ayres and Donohue refer to zero when they say the regression model would otherwise have predicted ) for at least the four years prior to the adoption of the law. For murder, the difference was statistically significant for only three to four years prior to adoption, but not in years one, two, five, or six. We will review our concerns with Ayres and Donohue s tests and how they interpret them in Part III, but even putting aside those concerns, it is relevant to point out that their own estimates provide a consistent story of the benefits from right-to-carry laws. Despite a nonrandom reporting of regressions that are not even consistent across tables, and using a hybrid model over a short period that overpredicts the costs, the vast majority of their evidence implies that the passage of concealed handgun laws reduces violent crime rates. E. The Estimated Benefits from the Law Table 1 takes all the county-level estimates reported by Ayres and Donohue in their current paper using the 1977 to 1997 data and applies their method of evaluating the changes in the social cost of crime from the concealed handgun law. Table 2 applies this method of determining the total costs or benefits to all the tables provided in Donohue s Brookings paper. 28 The separate estimates in Donohue s Brookings paper do include year-by-year effects of the law, in addition to the dummy, spline, and hybrid specifications. 29 Ayres and Donohue s estimated $1 billion increase in the annual costs of crime from the concealed handgun law relies on the hybrid estimates from their table The other two county hybrid results that they report imply annualized benefits from reduced crime of $1.7 and $1.05 billion. Despite their hybrid model overpredicting crime rates in early years, which is exacerbated by 28. Tables 10 and 11 in Ayres & Donohue, supra note 5, at 1262, 1269, and tables 8-1 and 8-3 in Donohue, supra note 7, at 294, 303, represent the same specifications. While these two sets of tables use identical specifications over the same 1977 to 1997 period, there are usually only small differences. The results are qualitatively the same. 29. A caveat worth noting is that Ayres and Donohue do not report the same sets of specifications (dummy, spline, and hybrid) across all tables. There is no discussion of why splines are reported in some tables and not others, but the omitted estimates frequently produce the largest benefits. In any case, we will stick to the sets of specifications that the authors report. 30. We calculate the estimated social costs/benefits for the dummy, spline, and hybrid models for the first five years of the law in the same way as Ayres and Donohue. However, Ayres and Donohue do not make these calculations for the regressions estimating the changing year-by-year impact of the law. Even though Ayres and Donohue include a prelaw crime rate trend (as used in the spline or hybrid models), no prelaw trend exists as a baseline for the year-by-year estimates. Therefore, we will use the crime rate when the law was passed. While this is the simplest approach, it also biases down (up) the gains (losses) from the law, especially for the aggravated assault when individual state-level trends are included (Figure 1d).

18 1330 STANFORD LAW REVIEW [Vol. 55:1313 the short five-year period they examine, their three hybrid estimates imply annualized benefits of $560 million. Even when the dummy variable estimates are included, which even Ayres and Donohue agree are flawed, their estimates imply an average annualized benefit of $233 million. Dropping the dummy estimates raises the estimated benefit to $1.34 billion per year. We also broke down Ayres and Donohue s state-level estimates reported in their tables 1 through 9 and their figures 3a through 3i. Using the same fiveyear period after the adoption of the law, there is an average annualized benefit from right-to-carry laws across all the specifications of $766 million. 31 Table 2 does the same breakdown for all the tables listed in Donohue s Brookings paper, using both the county- and state-level data from 1977 to While Donohue argues that this evidence strongly shows that concealed handgun laws are not beneficial, all but one of the estimates in his eight tables imply that the costs of crime fall with the passage of right-to-carry laws. The average estimate implies savings of $1.84 billion per year. Simple dummyvariable specifications imply much smaller annual gains from right-to-carry laws: They show a gain of $847 million using only the dummies versus an average benefit of $2.2 billion for the other specifications. Including a time trend for each individual state reduces the benefits estimated from county-level data from $2.1 billion to $233 million, though it produces a much smaller reduction in the estimated benefit for state-level data. At least for the first five years after the adoption of the law, the spline estimates imply benefits that are almost twice as large as those of the hybrids. The only estimate in Donohue s Brookings paper that implies that crime rates would rise uses only a postpassage dummy variable combined with individual state time trends. Returning to Table 1, the losses generated by Ayres and Donohue s table 13 are dominated by a few states. The table suggests that Kentucky s murder rate increased by an average of forty-two percent during the law s first five years, Louisiana s by thirty-four percent, and Tennessee s by thirty percent. Given that Ayres and Donohue estimated the five-year costs with only one full year of data for Kentucky, Louisiana, and South Carolina, some investigation seems warranted The year-by-year breakdown in the impact of the law reported in Figures 3a through 3i are the most detailed breakdown and they produce the largest benefit ($2.1 billion) of all the weighted least squares estimates. Among the two extremes for the other weighted least squares results, their dummy estimates imply an average loss per year of $354 million and their spline estimates imply an average benefit per year of $784 million. 32. All three states adopted the law in 1996, though few permits were issued in any of the states during the first year. Louisiana issued only 160 permits before Governor s Promises vs. Performance, ADVOCATE (Baton Rouge), Jan. 12, 1997, at 8A.

19 TABLE 1: COST-BENEFIT ANALYSES OF AYRES AND DONOHUE S COUNTY-LEVEL REGRESSIONS: FIVE-YEAR AVERAGE ESTIMATE OF NET COSTS/BENEFITS Model Total Murder Rape Robbery Assault Auto Theft Burglary Larceny Table 10 (4) Dummy # crimes , ,160 53,743 19, ,129 % crimes -7.7% -3.2% -0.3% -0.3% 10.8% 1.6% 9.4% $ cost/benefit -$1, $2, $ $4.7 -$29.3 $209.2 $28.2 $115.6 (5) Spline # crimes ,101-20,142-31,870-11,943-93, ,311 % crimes -8.1% -8.1% -10.8% -8.2% -2.4% -7.8% -3.3% $ cost/benefit -$3, $2, $ $ $ $46.5 -$ $40.6 (6) Hybrid # crimes ,646-16,226-16,630 16,421-59,769 6,322 % crimes -3.6% -4.3% -8.7% -4.3% 3.3% -5.0% 0.2% $ cost/benefit -$1, $ $ $ $419.9 $63.9 -$88.0 $2.5 Table 11: Includes state trends (3) Dummy # crimes ,459 17,417 5, ,438 % crimes -0.2% 2.6% 0.0% 7.1% 3.5% 0.5% 4.0% $ cost/benefit $ $54.4 $91.1 $0.0 $693.3 $67.8 $8.8 $49.2 (4) Hybrid # crimes ,885 30,166 1,493-44, ,633 % crimes -6.1% -1.5% -5.3% 7.8% 0.3% -3.7% 3.5% $ cost/benefit -$1, $1, $52.6 -$83.2 $761.6 $5.8 -$65.1 $43.0 Table 12: State-specific estimate, individual state time trends Dummy # crimes , , , , ,804.4 % crimes 0.9% 1.7% -0.9% 7.8% 4.7% 0.3% 4.2% $ cost/benefit $1,277.0 $237.7 $60.9 -$14.2 $795.8 $119.7 $8.4 $68.8 Table 13: State-specific estimate, individual state time trends Hybrid # crimes ,415 37,110 13, ,889 % crimes -0.3% 0.3% 0.3% 9.1% 5.7% 0.8% 4.6% $ cost/benefit $1, $93.9 $10.2 $4.3 $919.4 $144.4 $19.9 $74.6 Average predicted change # crimes ,182 3,450 6,590-6,951 36,587 % crimes -1.1% -0.6% -1.2% 0.9% 1.2% -0.6% 1.0% $ cost/benefit -$ $311 -$20 -$18 $87 $26 -$10 $14 Apr. 2003] CONFIRMING MORE GUNS, LESS CRIME 1331 PLASSMAN TABLE 1 4/16/2003 6:27 PM

20 TABLE 2: COST-BENEFIT ANALYSES OF COUNTY- AND STATE-LEVEL REGRESSIONS IN DONOHUE S BROOKINGS PAPER: FIVE-YEAR AVERAGE ESTIMATE OF NET COSTS/BENEFITS Model Total Murder Rape Robbery Assault Auto Theft Burglary Larceny Table 1: County data, the same as Ayres and Donohue s Table 10 (4) Dummy # crimes , ,825 17, ,562 % crimes -7.8% -2.9% -0.4% -0.1% 10.8% 1.5% 9.6% $ cost/benefit -$1, $2, $ $5.9 -$6.1 $209.5 $25.8 $117.7 (5) Spline # crimes -1,029-2,021-12,480 11,460-2,920-57,307 34,183 % crimes -11.7% -5.3% -6.7% 3.0% -0.6% -4.8% 1.1% $ cost/benefit -$3, $3, $ $105.0 $ $11.4 -$84.4 $13.3 (6) Hybrid # crimes ,697 40,781 11, ,286 % crimes -8.6% 0.1% -0.5% 10.5% 8.2% 0.9% 8.6% $ cost/benefit -$1, $2,341.8 $2.0 -$7.9 $1,027.5 $158.7 $16.4 $105.9 Table 2: State data (4) Dummy # crimes ,799-13,603-22,972 29,096-50,133 21,255 % crimes -4.5% -4.7% -7.3% -5.9% 5.8% -4.2% 0.7% $ cost/benefit -$2, $1, $ $ $580.0 $ $73.8 $8.3 (5) Spline # crimes ,585-17,437 5,329-26,443-52,117-9,245 % crimes -9.1% -4.1% -9.3% 1.4% -5.3% -4.4% -0.3% $ cost/benefit -$2, $2, $ $146.7 $ $ $76.7 -$3.6 (6) Hybrid # crimes ,593-16,932 2,953 8,504-39,656 46,642 % crimes -9.1% -4.2% -9.1% 0.8% 1.7% -3.3% 1.5% $ cost/benefit -$2, $2, $ $142.5 $74.5 $33.1 -$58.4 $18.1 Table 3: County data, Includes state trends, the same as Ayres and Donohue s Table 11 (4) Dummy # crimes 3 1, ,136 20,154 4, ,920 % crimes 0.0% 2.7% 0.3% 7.3% 4.1% 0.4% 4.1% $ cost/benefit $955.9 $8.5 $96.0 $5.1 $710.4 $78.4 $6.7 $50.9 (5) Hybrid # crimes ,406 30,404 4,451-44, ,267 % crimes -5.9% -1.5% -5.0% 7.9% 0.9% -3.7% 3.7% $ cost/benefit -$ $1, $51.3 -$79.2 $767.6 $17.3 -$65.2 $45.6 continued 1332 STANFORD LAW REVIEW [Vol. 55:1313 PLASSMAN TABLE 2, PAGE 1 4/16/2003 6:27 PM

21 TABLE 2 (CONTINUED) Model Total Murder Rape Robbery Assault Auto Theft Burglary Larceny Table 4: State data, includes state trends (3) Dummy # crimes , ,720 7,890 63,567 % crimes -1.9% -1.2% -0.7% 0.1% 4.6% 0.7% 2.0% $ cost/benefit -$ $ $40.3 -$10.2 $12.2 $88.4 $11.6 $24.7 (4) Hybrid # crimes ,599-18,333 6,304-14,087-31,292 29,805 % crimes -7.9% -4.2% -9.8% 1.6% -2.8% -2.6% 0.9% $ cost/benefit -$2, $2, $ $154.3 $ $54.8 -$46.1 $11.6 Table 5: County data Year # crimes ,921-8,560-15,818 9,256-20, ,691 dummies % crimes -5.8% -7.6% -4.6% -4.1% 1.9% -1.7% 5.4% $ cost/benefit -$2, $1, $ $72.0 -$399.4 $36.0 -$29.6 $66.4 Table 6: State Data Year # crimes ,374-17,811-19,492-11,346-79,493-5,058 dummies % crimes -6.7% -6.2% -9.6% -5.0% -2.3% -6.7% -0.2% $ cost/benefit -$2, $1, $ $ $ $44.2 -$ $2.0 Table 7: County data, includes state trends Year # crimes ,106-5,856 36,238-9,554-64, ,272 dummies % crimes -5.2% -2.9% -3.1% 9.4% -1.9% -5.4% 7.2% $ cost/benefit -$ $1, $ $49.3 $ $37.2 -$95.5 $88.4 Table 8: State data, includes state trends Year # crimes ,032-32,116 7,890-56, ,214-90,087 dummies % crimes -9.6% -7.9% -17.2% 2.0% -11.3% -9.2% -2.9% $ cost/benefit Average predicted change # crimes ,363-11,056 7,955 4,868-36,314 93,719 % crimes -6.7% -3.6% -5.9% 2.1% 1.0% -3.0% 3.0% $ cost/benefit Average $cost/benefit predicted change by type Dummy Spline -$3, $2, $ $125.9 $ $57.2 -$80.6 $ 4.9 Hybrid -$1, $2, $85.4 -$96.0 $ $ $38.3 $ 45.3 Year Dummies -$2, $1, $ $135.3 $ $66.0 -$101.3 $ 29.2 Apr. 2003] CONFIRMING MORE GUNS, LESS CRIME 1333 PLASSMAN TABLE 2, PAGE 2 4/16/2003 6:27 PM

22 1334 STANFORD LAW REVIEW [Vol. 55:1313 Although a forty-two percent increase in Kentucky s murder rate should be easily spotted, this is not the case. Figure 3 shows the actual change in Kentucky s murder rates during the 1990s, and compares it to the change in murder rates for other states in the Midwest and for the United States as a whole. While the United States and Midwest murder rates were either unchanged or falling from 1992 to 1995, Kentucky s murder rate was rising. Kentucky s murder rate fell when the law was just getting started in 1996, and continued declining after that. Both percentage declines were much greater than the declines over the same period for the rest of the Midwest or the United States as a whole. Nor do other factors imply that Kentucky should have had an even bigger drop had it not been for the detrimental impact of the law (for example, Kentucky s arrest rate declined by forty percent between 1995 and 1998, and continued declining after that) A similar breakdown for Louisiana, South Carolina, and Tennessee is available from the authors.

23 Murder rate FIGURE 3: EXAMINING THE LAST STATES TO ADOPT THE RIGHT-TO-CARRY LAW Year Kentucky Midwest United States Apr. 2003] CONFIRMING MORE GUNS, LESS CRIME 1335 PLASSMAN FIGURE 3 4/16/2003 6:26 PM

24 1336 STANFORD LAW REVIEW [Vol. 55:1313 II. COUNTY-LEVEL DATA FROM 1977 TO 2000 A. Advantages and Disadvantages of Different Types of Data While most crime analysis has traditionally been done at the state level, disaggregated data has an important advantage in that both crime and arrest rates vary widely within states. In fact, the variation in both crime and arrest rates across states is almost always smaller than the average within-state variation across counties. 34 It is no more accurate to view all counties in the typical state as a homogenous unit than it is to view all states in the United States as one homogenous unit. For example, when a state s arrest rate rises, it may make a big difference whether that increase is taking place in the most or least crime-prone counties, or whether it is increasing a lot in one jurisdiction or across the entire state. A simple example can show this potential aggregation bias. Assume, for the sake of argument, that income is negatively correlated with property crimes (that is, a person with higher income commits fewer property crimes). Assume that Location A has a population of 1000 persons and Location B has a population of 2000 persons, with respective per capita incomes of $50,000 and $40,000 and property crimes of 100 and 200. Now assume that the per capita income at Location A increases to $60,000 and crimes fall to eighty, and the per capita income at Location B decreases to $36,000 and the number of crimes increases to 240. If we examined both locations separately, we would detect that increases in income lead to fewer property crimes and vice versa. However, if we instead aggregated both locations into a single location, we would observe that overall per capita income had increased (from $43,334 to $44,000) and that the number of property crimes committed had increased as well (from 300 to 320). Aggregating the data gives the false impression that increases in income lead to increases in the number of property crimes. While there is a fair degree of similarity between state- and county-level data, as shown by Ayres and Donohue s yearly breakdown of the impact of right-to-carry laws, we will concentrate here on the county-level data, both because we believe that it provides a much more accurate measure of changes in crime rates and because of time and space constraints. B. Extending the Data to 2000 There are seven different types of estimates that have been used to evaluate the impact of right-to-carry laws: a dummy variable for the law; before-and- 34. LOTT, supra note 6, at

25 Apr. 2003] CONFIRMING MORE GUNS, LESS CRIME 1337 after trends; a hybrid approach; the impact of the law on a year-by-year basis before and after the law; nonlinear before-and-after trends; some county-level data on the per capita number of permits issued; and the predicted number of permits based upon the characteristics of the right-to-carry laws and some limited information on permit issuance in some states. We analyze the county-level data from 1977 to 2000 using the four different methods discussed in Ayres and Donohue s paper and in Donohue s Brookings paper. We provide a more precise breakdown of year-by-year impacts from the law in the Appendix, but for the text we follow the two-year interval approach used by Donohue so as to make our results more comparable. The regressions use all control variables employed in the second edition of More Guns, Less Crime. 35 In addition to arrest rates, we use: per capita income; population density; the execution rate for murder; per capita welfare payments; per capita unemployment insurance payments; retirement payments per person over age sixty-five; thirty-six different demographic measures by age, gender, and race; state-level poverty and unemployment rates; county and regional year-fixed effects. 36 Figures 4a through 4f graphically report the results for a postpassage dummy, spline, hybrid, and year-by-year impacts. 37 Two conclusions are readily apparent from these Figures. First, the year-by-year impacts of the law are very similar to those reported by Ayres and Donohue for the 1977 to 1997 period. The first six years during which the right-to-carry law is in effect are associated with about ten percent declines in murder and rape and an eight percent decline in robbery rates. The year-by-year breakdown in the Appendix shows that by the second full year of the law, all four violent crime categories have experienced large drops, with murder falling by 5% and robbery by 8.7% Lott, supra note Despite Ayres and Donohue s claim that many of the explanatory variables are only measured on the state level, Ayres & Donohue, supra note 5, at 1259, with the exception of the state-level poverty and unemployment rates these are all county-level control variables. The regressions reported in this Response use crime-specific arrest rates. We repeated the analyses of murder, rape, robbery, and aggravated assault with the arrest rate of violent crime and obtained very similar estimates. 37. The other property crime rates are not included to save space, but are available from the authors. 38. For rapes and aggravated assaults, the small increases from year zero to year one seem at least in part to represent an upward trend that had been occurring for those crime rates over a period of eight years in the case of rape and five years for aggravated assault.